摘要：Next generation sequencing (NGS) technologies have greatly enhanced our ability to identify new viral pathogens in various types of biological samples. This approach has led to the discovery of new viruses and has revealed hidden associations of viromes with many diseases. However, unlike the 16S rRNA, which allows for bacterial detection by metabarcoding, the diversity and variability of viral genomes render the creation of universal oligonucleotides for targeting all known and novel viruses impossible. While wholegenome sequencing solves this problem, its efficiency is inadequate due to the high cost per sample and relatively low sensitivity. Furthermore, the existing approaches to designing oligonucleotides for targeted PCR enrichment are usually incomprehensive, being oriented at detecting a particular viral species or a genus based on the presumption of its presence in the sample. In this study, we developed a computational pipeline for designing genus-specific oligonucleotides that would simultaneously cover a multitude of known viruses from different taxonomic groups. This new tool was used to design an oligonucleotide panel for targeted enrichment of viral nucleic acids in different types of samples, and its applicability for the detection of multiple viral genera at once was demonstrated. Next, we created a custom protocol for NGS library preparation adapted to the new primer panel, which was tested together on a number of samples and proved highly efficient in pathogen detection and identification. Since a reliable algorithm for bioinformatic analysis is crucial for rapid classification of the sequences, in this work, we developed an NGSbased data analysis module and demonstrated its functionality both for detecting novel viruses and analyzing virome diversity. This work was supported by an RSF (Russian Science Foundation) grant (No. 17-74-20096).
